83704 2012-04-11 10:34:48 -0700 [GTK] media/event-attributes.html fails if mixer is not at 100% 2021-08-12 15:52:46 -0700 1 1 1 Unclassified WebKit Tools / Tests 528+ (Nightly build) Unspecified Unspecified RESOLVED FIXED https://bugs.webkit.org/show_bug.cgi?id=229057 Gtk, InRadar, LayoutTestFailure P2 Normal --- 131545 158914 201266 1 gustavo calvaris abarth bugs-noreply calvaris dpranke d-r jussi.kukkonen mcatanzaro naginenis pnormand rakuco spena tmpsantos webkit-bug-importer webkit.review.bot zan oldest_to_newest 600378 0 gustavo 2012-04-11 10:34:48 -0700 This is not that problematic for our bots, since we can control that, but it means the test won't pass locally in some cases, which is bad. 601037 1 spena 2012-04-12 02:36:13 -0700 I guess we could probably use amixer: <http://linux.die.net/man/1/amixer>, the same way we're using pactl, and set the master volume to 100% before running the tests, and restoring it afterwards. 602000 2 137073 spena 2012-04-13 06:01:30 -0700 Created attachment 137073 Patch 602004 3 spena 2012-04-13 06:06:49 -0700 (In reply to comment #2) > Created an attachment (id=137073) [details] > Patch It's worth noting that I've been unable to reproduce the claimed bug :-/. In my system, with the mixer either muted or at non-100% volume, it doesn't result on that test failing. Before applying, it would be worth trying if this patch is really helping in your case, Gustavo. Also, we would need to state in the wiki at <https://trac.webkit.org/wiki/WebKitGtkLayoutTests> that alsa-utils are now also recommended. Let's not hurry pushing this, since Phil also said this test was failing in his environment, so he can confirm whether this helps or not. 602025 4 pnormand 2012-04-13 06:44:50 -0700 In my Debian setup alsactl is in /usr/sbin which is not in my PATH. This needs to be double-checked before deployment of this patch, the GTK bots run on Debian. 602028 5 pnormand 2012-04-13 06:47:27 -0700 Hum even with the patch and my PATH fixed I get those 2 failing :( media/event-attributes.html = TEXT media/sources-fallback-codecs.html = TEXT 606086 6 137073 pnormand 2012-04-19 16:12:09 -0700 Comment on attachment 137073 Patch Marking r- as it doesn't seem to totally fix the issue, although I liked this approach :/ 623071 7 rakuco 2012-05-14 10:16:49 -0700 What does the failure look like? We've been experiencing some failures on the EFL bot for a few revisions in this test. 623088 8 pnormand 2012-05-14 10:32:36 -0700 (In reply to comment #7) > What does the failure look like? We've been experiencing some failures on the EFL bot for a few revisions in this test. Volume value mismatches. 623096 9 rakuco 2012-05-14 10:41:05 -0700 Like this? @@ -13,14 +13,14 @@ *** changing playback rate RUN(video.playbackRate = 2) +EVENT(volumechange) + +*** pausing playback +RUN(video.pause()) EVENT(ratechange) *** setting volume RUN(video.volume = 0.5) -EVENT(volumechange) - -*** pausing playback -RUN(video.pause()) EVENT(pause) *** seeking 623099 10 pnormand 2012-05-14 10:44:35 -0700 Yep! 623102 11 rakuco 2012-05-14 10:46:01 -0700 Hmm, both ports seem to be affected then, thanks for the answers. Retitling the bug. 716040 12 jussi.kukkonen 2012-09-10 06:59:57 -0700 I think there's an actual bug here: unless the system volume is 100% there's an extra volumechange event (as can be seen in rakucos diff). Originally the volume is 1.0. Then, apparently after play(), the volume is changed to the actual system volume (triggering the volumechange event if the system volume is something else than 100%). If I was a html app developer I'd find this very surprising. Others, (chromium, firefox) do not expose the system volume through HTMLMediaElement.volume: HTMLMediaElement.volume 1.0 effectively means "the current system volume" and changing the system volume does not trigger volumechange events Now, a case could be made for exposing system volume in the HTMLMediaElement (spec says "user agents may remember the last set value across sessions, on a per-site basis or otherwise, so the volume may start at other values"): in that case the original value "1.0" was wrong, it should always start at system volume. 716054 13 163124 jussi.kukkonen 2012-09-10 07:14:55 -0700 Created attachment 163124 test volumechange event test that shows the extra volumechange event (if system volume is < 100%). Output on my machine, EFL port: --- Log: * canplaythrough event, mute is false, volume is 1. Now calling play() * volumechange event, mute is false, volume is 0.4220733642578125 --- 716145 14 pnormand 2012-09-10 09:15:03 -0700 (In reply to comment #12) > I think there's an actual bug here: unless the system volume is 100% there's an extra volumechange event (as can be seen in rakucos diff). > > Originally the volume is 1.0. Then, apparently after play(), the volume is changed to the actual system volume (triggering the volumechange event if the system volume is something else than 100%). If I was a html app developer I'd find this very surprising. > Yes, this is due to PulseAudio's flat volumes feature combined with its modules to restore stream and device volume values I'm afraid. I think we get an extra volumechange event because of that. Before running your test case can you disable the module-device-restore and module-stream-restore modules? And also disable the Flat volumes thing? Speaking of a solution... maybe we should be more careful when monitoring the playbin2 volume property. Not sure what to do though, ideas welcome! 716248 15 jussi.kukkonen 2012-09-10 10:43:25 -0700 (In reply to comment #14) > Yes, this is due to PulseAudio's flat volumes feature combined with its modules to restore stream and device volume values I'm afraid. > > I think we get an extra volumechange event because of that. Before running your test case can you disable the module-device-restore and module-stream-restore modules? And also disable the Flat volumes thing? I still get the volumechange. but I think I failed to disable everything as pulseaudio clearly restores the volume for something (the element always gets the volume that was last set to a mediaelement). But you are right, it's not system volume, it's definitely "app-specific". > Speaking of a solution... maybe we should be more careful when monitoring the playbin2 volume property. Not sure what to do though, ideas welcome! Originally I thought we should just do what everyone else does (let HTMLElement volume=1.0 represent "current master volume")... but now I'd rather we fixed whatever goes wrong here and have pulseaudio handle this, it really does feel like a well working solution with Flat volumes and as far as I understand should be compliant as well. So I think we just need to initialize the MediaElement properties as early as possible. It's not acceptable that muted and volume change drastically _after_ the player app calls MediaElement.play(). It's been a while since I've done anything with gstreamer so I'll have to read up on that... 717028 16 jussi.kukkonen 2012-09-11 03:44:26 -0700 (In reply to comment #14) > Yes, this is due to PulseAudio's flat volumes feature combined with its modules to restore stream and device volume values I'm afraid. I've now re-tested, and it happens without any "-restore" modules (just flat volume feature seems to be enough). Without flat volume the htmlelement volume stayed at 1.0 (In reply to comment #15) > So I think we just need to initialize the MediaElement properties as early > as possible. It's not acceptable that muted and volume change drastically > _after_ the player app calls MediaElement.play(). Based on looking at playbin2 and asking on #gstreamer, it seems this is the best we can get... playbin2 cannot know the volume it will have until it's playing. Also, "muted" property does not actually change as I claimed, it's just "volume" that is adjusted. 721822 17 jussi.kukkonen 2012-09-17 05:38:07 -0700 Oh, there's a bug in the patch: it sets the volume to "100" (without units): the result depends on the hardware (on my HW it ends up as <0.01%). should be "100%". Patch WFM after fixing that -- although it's a nasty hack: setting system volume to 100% even for a short time is just wrong. Also, documenting what I've learned on this for the benefit of anyone else trying to figure it out: * This phenomenon of extra volume events is a feature of Pulseaudio as Philippe says (flat volumes + stream volume restore): we cannot tell beforehand what the volume of a stream will be. * It seems to be spec compliant * We as a pulseaudio using app cannot really avoid this even if we wanted to (at least I have not figured out how that is possible). The volume we show in UI and JS has to be the playbin volume (which usually, but not necessarily, is a PA stream volume). * The additional volumechange and mutechange events are _unavoidable_, but we can make them happen in a more reasonable point in time: the data is available when we reach GST_STATE_PAUSED (instead of getting them after playback starts). This is a different bug though, I'll file that. 721825 18 pnormand 2012-09-17 05:44:25 -0700 (In reply to comment #17) > Oh, there's a bug in the patch: it sets the volume to "100" (without units): the result depends on the hardware (on my HW it ends up as <0.01%). should be "100%". > > Patch WFM after fixing that -- although it's a nasty hack: setting system volume to 100% even for a short time is just wrong. > > Well this is for tests purpose. > * The additional volumechange and mutechange events are _unavoidable_, but we can make them happen in a more reasonable point in time: the data is available when we reach GST_STATE_PAUSED (instead of getting them after playback starts). This is a different bug though, I'll file that. Please CC me :) 723384 19 jussi.kukkonen 2012-09-19 01:28:10 -0700 I came back to thinking about this because we really should have unskipped tests before poking things like bug 96912. I'm not very familiar with the test system itself but... would it be ok to compile a PA-using app that modifies the saved stream volume for this process (based on the name "WebProcess" or whatever) before starting the test (so we could make sure the volume is 100% or another value), and then puts it back to what it was after testing? This would have the least effect on the system: no "system volume" changes and no changes to already running WebProcesses: only WebProcesses started after running this setup-app would be affected. The code itself is not complex (couple of call to pulse/ext-stream-restore.h and the logic to save the old values) and I believe it would replace both the amixer trick in the patch and PulseAudioSanitizer class already in the repo. Unfortunately there's no python api for this and doing this with ctypes would require more knowledge than I have... So, is adding compiled test-setup-apps that depend on libpulse a bad idea? If it is then let's just get that amixer patch committed. 724371 20 jussi.kukkonen 2012-09-20 03:35:27 -0700 (In reply to comment #19) > would it be ok to compile a PA-using app... philn had some understandable reservations on exactly this and pointed me to pulseaudio module-match, which does look like it might do the trick. Unfortunately the only thing I can do with module-match is crash the daemon: https://bugs.freedesktop.org/show_bug.cgi?id=55135. At this point this looks like a pulseaudio bug. 780025 21 jussi.kukkonen 2012-11-30 05:15:16 -0800 So, the pulseaudio module approach might work, once we can rely on the crash fix to be there... But that might take a long while. In the meantime should we go with the brute force method as in Simons patch for now -- forcing the system volume to 100% at test startup either using alsautils like here or with pulseaudio scripts (I assume that has to be possible)? (In reply to comment #5) > Hum even with the patch and my PATH fixed I get those 2 failing :( > media/event-attributes.html = TEXT > media/sources-fallback-codecs.html = TEXT I'm guessing this is because of the typo, "100" instead of "100%" 781319 22 pnormand 2012-12-03 04:01:44 -0800 Which typo? When calling amixer? 781521 23 jussi.kukkonen 2012-12-03 09:11:19 -0800 (In reply to comment #22) > Which typo? When calling amixer? Yes, mentioned it in comment 17. Should be: amixer_process = subprocess.Popen(["amixer", "set", "Master", "100%", "unmute"], stderr=devnull, stdout=devnull) otherwise the value will be handled as hardware value. 781551 24 pnormand 2012-12-03 09:46:32 -0800 Oh right I forgot about comment 17 :) 1750140 25 pnormand 2021-04-14 08:58:19 -0700 This is no longer an issue because the test harness relies on fakeaudiosink. 1750163 26 calvaris 2021-04-14 09:35:15 -0700 Even with fakeaudiosink, I see some extra volume events, that need study. Test still fails. 1773765 27 432601 calvaris 2021-06-30 09:01:11 -0700 Created attachment 432601 Patch 1773766 28 432602 calvaris 2021-06-30 09:05:08 -0700 Created attachment 432602 Patch 1774067 29 432602 pnormand 2021-07-01 04:04:59 -0700 Comment on attachment 432602 Patch View in context: https://bugs.webkit.org/attachment.cgi?id=432602&action=review > Tools/WebKitTestRunner/TestController.cpp:990 > + // GStreamer uses fakesink to avoid sound output during testing and doing this creates trouble with volume events. > +#if !USE(GSTREAMER) > WKPageSetMediaVolume(m_mainWebView->page(), 0); > +#endif That feels like closing our eyes on a real bug though :( fakeaudiosink is used as a mock renderer, it should however handle volume/mute properties reliably. So I wonder if there's a real issue in our code handling volume events. 1774102 30 calvaris 2021-07-01 07:15:00 -0700 (In reply to Philippe Normand from comment #29) > Comment on attachment 432602 [details] > Patch > > View in context: > https://bugs.webkit.org/attachment.cgi?id=432602&action=review > > > Tools/WebKitTestRunner/TestController.cpp:990 > > + // GStreamer uses fakesink to avoid sound output during testing and doing this creates trouble with volume events. > > +#if !USE(GSTREAMER) > > WKPageSetMediaVolume(m_mainWebView->page(), 0); > > +#endif > > That feels like closing our eyes on a real bug though :( fakeaudiosink is > used as a mock renderer, it should however handle volume/mute properties > reliably. So I wonder if there's a real issue in our code handling volume > events. I don't know what others port to regarding volume management but we do rely on what GStreamer is telling us. If we set page volume to 0, we eventually set GStreamer volume to 0, which ends up triggering volume changes. It is not closing our eyes on a real bug, it is a different way of mocking among different ports. We have fakeaudiosink that does everything that a real renderer does but without sending anything to the sound card and it also handles the volume changes as expected. Actually, while testing I am going to unflag several other bugs that were failing because of this. 1774111 31 432694 calvaris 2021-07-01 08:08:17 -0700 Created attachment 432694 Patch 1774112 32 calvaris 2021-07-01 08:10:07 -0700 If this patch lands, I'll close the related bugs for the expectations I removed. 1774373 33 ews-feeder 2021-07-02 00:01:28 -0700 Committed r279499 (239351@main): <https://commits.webkit.org/239351@main> All reviewed patches have been landed. Closing bug and clearing flags on attachment 432694. 1774374 34 webkit-bug-importer 2021-07-02 00:02:17 -0700 <rdar://problem/80067853> 137073 2012-04-13 06:01:30 -0700 2021-06-30 09:00:55 -0700 Patch bug-83704-20120413150120.patch text/plain 4756 spena U3VidmVyc2lvbiBSZXZpc2lvbjogMTEzOTgxCmRpZmYgLS1naXQgYS9Ub29scy9DaGFuZ2VMb2cg Yi9Ub29scy9DaGFuZ2VMb2cKaW5kZXggYTM5YmFkZDMwNDdlM2QyZTdkZWQ4ZGVlMDdkNjMwOTE1 YjIyYzUwZS4uZGNmZjUzZTAwYjNjZDhhMGE0ZDliYmYyYTIyNjY2Yjg5NDY5Y2M0NiAxMDA2NDQK LS0tIGEvVG9vbHMvQ2hhbmdlTG9nCisrKyBiL1Rvb2xzL0NoYW5nZUxvZwpAQCAtMSwzICsxLDI1 IEBACisyMDEyLTA0LTEzICBTaW1vbiBQZW5hICA8c3BlbmFAaWdhbGlhLmNvbT4KKworICAgICAg ICBbR1RLXSBtZWRpYS9ldmVudC1hdHRyaWJ1dGVzLmh0bWwgZmFpbHMgaWYgbWl4ZXIgaXMgbm90 IGF0IDEwMCUKKyAgICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lk PTgzNzA0CisKKyAgICAgICAgUmV2aWV3ZWQgYnkgTk9CT0RZIChPT1BTISkuCisKKyAgICAgICAg V2hlbiBydW5uaW5nIHRlc3RzLCBzZXQgdGhlIG1peGVyIHZvbHVtZSB0byAxMDAlLiBCZWZvcmUg dGhlIHRlc3RzCisgICAgICAgIGFyZSBydW4sIHNhdmUgdGhlIGN1cnJlbnQgYWxzYW1peGVyIHN0 YXRlLCBzbyBpdCBpcyByZXN0b3JlZAorICAgICAgICBhZnRlcndhcmRzLiBJZiBhbHNhLXV0aWxz IGFyZSBub3QgZm91bmQsIHRoaXMgcGF0Y2ggd29uJ3QgbW9kaWZ5IHRoZQorICAgICAgICBtaXhl ciB2b2x1bWUuCisKKyAgICAgICAgKiBTY3JpcHRzL3dlYmtpdHB5L2xheW91dF90ZXN0cy9wb3J0 L2d0ay5weToKKyAgICAgICAgKEd0a1BvcnQuX3NhdmVfYW1peGVyX3N0YXRlKTogU2F2ZXMgYWxz YW1peGVyIGN1cnJlbnQgc3RhdGUgYW5kIHNldHMKKyAgICAgICAgdGhlIG1peGVyIHZvbHVtZSB0 byAxMDAlCisgICAgICAgIChHdGtQb3J0Ll9yZXN0b3JlX2FtaXhlcl9zdGF0ZSk6IFJlc3RvcmVz IHRoZSBzdGF0ZSB0byBpdHMgb3JpZ2luYWwKKyAgICAgICAgdmFsdWUuCisgICAgICAgIChHdGtQ b3J0LnNldHVwX3Rlc3RfcnVuKTogQ2FsbHMgdGhlIG1ldGhvZCB0byBzYXZlIHRoZSBzdGF0ZQor ICAgICAgICBiZWZvcmUgdGhlIHRlc3RzIGFyZSBydW4uCisgICAgICAgIChHdGtQb3J0LmNsZWFu X3VwX3Rlc3RfcnVuKTogQ2FsbHMgdGhlIG1ldGhvZCB0byByZXN0b3JlIHRoZSBzdGF0ZQorICAg ICAgICBvbmNlIHRoZSB0ZXN0cyBmaW5pc2hlZC4KKwogMjAxMi0wNC0xMiAgU2ltb24gUGVuYSAg PHNwZW5hQGlnYWxpYS5jb20+CiAKICAgICAgICAgW0dUS10gVGVzdCBydW5uZXIgY3Jhc2hlcyB3 aXRob3V0IHJlY2VudCB2ZXJzaW9ucyBvZiBwdWxzZWF1ZGlvLXV0aWxzCmRpZmYgLS1naXQgYS9U b29scy9TY3JpcHRzL3dlYmtpdHB5L2xheW91dF90ZXN0cy9wb3J0L2d0ay5weSBiL1Rvb2xzL1Nj cmlwdHMvd2Via2l0cHkvbGF5b3V0X3Rlc3RzL3BvcnQvZ3RrLnB5CmluZGV4IDcxZTVhM2I3OTIw NzFjNmM5NTM1MzNjYzg4MDg0M2EwNjRjMmRmMzguLjhhNzgyYTIxMDUwMzIwZTAxYjdhNTI3MzNk YTkzZjA2ODcxZGYwYmUgMTAwNjQ0Ci0tLSBhL1Rvb2xzL1NjcmlwdHMvd2Via2l0cHkvbGF5b3V0 X3Rlc3RzL3BvcnQvZ3RrLnB5CisrKyBiL1Rvb2xzL1NjcmlwdHMvd2Via2l0cHkvbGF5b3V0X3Rl c3RzL3BvcnQvZ3RrLnB5CkBAIC0zMCw2ICszMCw3IEBAIGltcG9ydCBsb2dnaW5nCiBpbXBvcnQg b3MKIGltcG9ydCBzaWduYWwKIGltcG9ydCBzdWJwcm9jZXNzCitpbXBvcnQgdGVtcGZpbGUKIAog ZnJvbSB3ZWJraXRweS5sYXlvdXRfdGVzdHMubW9kZWxzLnRlc3RfY29uZmlndXJhdGlvbiBpbXBv cnQgVGVzdENvbmZpZ3VyYXRpb24KIGZyb20gd2Via2l0cHkubGF5b3V0X3Rlc3RzLnBvcnQuc2Vy dmVyX3Byb2Nlc3MgaW1wb3J0IFNlcnZlclByb2Nlc3MKQEAgLTExMCw2ICsxMTEsMzIgQEAgY2xh c3MgR3RrUG9ydChXZWJLaXRQb3J0KToKICAgICAgICAgICAgICAgICAgICAgICAgIF9sb2cuZGVi dWcoJ1VuYWJsZSB0byBwYXJzZSBtb2R1bGUgaW5kZXguIFBsZWFzZSBjaGVjayBpZiB5b3VyIHB1 bHNlYXVkaW8tdXRpbHMgdmVyc2lvbiBpcyB0b28gb2xkLicpCiAgICAgICAgICAgICAgICAgcmV0 dXJuCiAKKyAgICBkZWYgX3NhdmVfYW1peGVyX3N0YXRlKHNlbGYpOgorICAgICAgICB3aXRoIG9w ZW4ob3MuZGV2bnVsbCwgJ3cnKSBhcyBkZXZudWxsOgorICAgICAgICAgICAgc2VsZi5fYW1peGVy X3N0YXRlID0gdGVtcGZpbGUubWtzdGVtcCgpWzFdCisgICAgICAgICAgICAjIFNhdmluZyBhbHNh bWl4ZXIgc3RhdGUgYmVmb3JlIHJ1bm5pbmcgdGhlIHRlc3RzCisgICAgICAgICAgICB0cnk6Cisg ICAgICAgICAgICAgICAgYWxzYWN0bF9wcm9jZXNzID0gc3VicHJvY2Vzcy5Qb3BlbihbImFsc2Fj dGwiLCAic3RvcmUiLCAiLWYiLCBzZWxmLl9hbWl4ZXJfc3RhdGVdLCBzdGRlcnI9ZGV2bnVsbCwg c3Rkb3V0PWRldm51bGwpCisgICAgICAgICAgICAgICAgYWxzYWN0bF9wcm9jZXNzLndhaXQoKQor ICAgICAgICAgICAgICAgIGlmIGFsc2FjdGxfcHJvY2Vzcy5yZXR1cm5jb2RlID09IDA6CisgICAg ICAgICAgICAgICAgICAgIF9sb2cuZGVidWcoJ2Fsc2FtaXhlciBzdGF0ZSBzdG9yZWQgc3VjY2Vz c2Z1bGx5JykKKyAgICAgICAgICAgICAgICBlbHNlOgorICAgICAgICAgICAgICAgICAgICBfbG9n LmRlYnVnKCdVbmFibGUgdG8gc3RvcmUgYWxzYW1peGVyIHN0YXRlJykKKyAgICAgICAgICAgIGV4 Y2VwdCBPU0Vycm9yOgorICAgICAgICAgICAgICAgIF9sb2cuZGVidWcoJ2Fsc2FjdGwgbm90IGZv dW5kLiBQbGVhc2UgaW5zdGFsbCBhbHNhLXV0aWxzIHRvIGF2b2lkIHBvdGVudGlhbCBtZWRpYSB0 ZXN0IGZhaWx1cmVzJykKKyAgICAgICAgICAgICAgICBvcy5yZW1vdmUoc2VsZi5fYW1peGVyX3N0 YXRlKQorICAgICAgICAgICAgICAgIHJldHVybgorICAgICAgICAgICAgdHJ5OgorICAgICAgICAg ICAgICAgICMgU2V0dGluZyBNYXN0ZXIgdm9sdW1lIHRvIDEwMCUgd2hpbGUgdGhlIHRlc3RzIGFy ZSBydW4KKyAgICAgICAgICAgICAgICBhbWl4ZXJfcHJvY2VzcyA9IHN1YnByb2Nlc3MuUG9wZW4o WyJhbWl4ZXIiLCAic2V0IiwgIk1hc3RlciIsICIxMDAiLCAidW5tdXRlIl0sIHN0ZGVycj1kZXZu dWxsLCBzdGRvdXQ9ZGV2bnVsbCkKKyAgICAgICAgICAgICAgICBhbWl4ZXJfcHJvY2Vzcy53YWl0 KCkKKyAgICAgICAgICAgICAgICBpZiBhbWl4ZXJfcHJvY2Vzcy5yZXR1cm5jb2RlID09IDA6Cisg ICAgICAgICAgICAgICAgICAgIF9sb2cuZGVidWcoJ1N1Y2Nlc3NmdWxseSBzZXQgTWFzdGVyIHZv bHVtZSB0byAxMDAlJykKKyAgICAgICAgICAgICAgICBlbHNlOgorICAgICAgICAgICAgICAgICAg ICBfbG9nLmRlYnVnKCdVbmFibGUgdG8gc2V0IE1hc3RlciB2b2x1bWUgdG8gMTAwJScpCisgICAg ICAgICAgICBleGNlcHQgT1NFcnJvcjoKKyAgICAgICAgICAgICAgICBfbG9nLmRlYnVnKCdhbWl4 ZXIgbm90IGZvdW5kLiBQbGVhc2UgaW5zdGFsbCBhbHNhLXV0aWxzIHRvIGF2b2lkIHBvdGVudGlh bCBtZWRpYSB0ZXN0IGZhaWx1cmVzJykKKwogICAgIGRlZiBfcmVzdG9yZV9wdWxzZWF1ZGlvX21v ZHVsZShzZWxmKToKICAgICAgICAgIyBJZiBwdWxzZWF1ZGlvJ3MgbW9kdWxlLXN0cmVhbS1yZXN0 b3JlIHdhcyBwcmV2aW91c2x5IHVubG9hZGVkLAogICAgICAgICAjIHJlc3RvcmUgaXQgYmFjay4g V2Ugc2hvdWxkbid0IG5lZWQgZXh0cmEgY2hlY2tzIGhlcmUsIHNpbmNlIGFuCkBAIC0xMjMsMTIg KzE1MCwyNCBAQCBjbGFzcyBHdGtQb3J0KFdlYktpdFBvcnQpOgogICAgICAgICAgICAgICAgIGVs c2U6CiAgICAgICAgICAgICAgICAgICAgIF9sb2cuZGVidWcoJ1Jlc3RvcmluZyBtb2R1bGUtc3Ry ZWFtLXJlc3RvcmUgZmFpbGVkJykKIAorICAgIGRlZiBfcmVzdG9yZV9hbWl4ZXJfc3RhdGUoc2Vs Zik6CisgICAgICAgIGlmIHNlbGYuX2FtaXhlcl9zdGF0ZToKKyAgICAgICAgICAgIHdpdGggb3Bl bihvcy5kZXZudWxsLCAndycpIGFzIGRldm51bGw6CisgICAgICAgICAgICAgICAgYWxzYWN0bF9w cm9jZXNzID0gc3VicHJvY2Vzcy5Qb3BlbihbImFsc2FjdGwiLCAicmVzdG9yZSIsICItZiIsIHNl bGYuX2FtaXhlcl9zdGF0ZV0sIHN0ZGVycj1kZXZudWxsLCBzdGRvdXQ9ZGV2bnVsbCkKKyAgICAg ICAgICAgICAgICBhbHNhY3RsX3Byb2Nlc3Mud2FpdCgpCisgICAgICAgICAgICAgICAgaWYgYWxz YWN0bF9wcm9jZXNzLnJldHVybmNvZGUgPT0gMDoKKyAgICAgICAgICAgICAgICAgICAgX2xvZy5k ZWJ1ZygnUmVzdG9yZWQgYWxzYW1peGVyIHN0YXRlJykKKyAgICAgICAgICAgICAgICBlbHNlOgor ICAgICAgICAgICAgICAgICAgICBfbG9nLmRlYnVnKCdVbmFibGUgdG8gcmVzdG9yZSBhbHNhbWl4 ZXIgc3RhdGUnKQorICAgICAgICAgICAgICAgIG9zLnJlbW92ZShzZWxmLl9hbWl4ZXJfc3RhdGUp CiAKICAgICBkZWYgc2V0dXBfdGVzdF9ydW4oc2VsZik6CiAgICAgICAgIHNlbGYuX3VubG9hZF9w dWxzZWF1ZGlvX21vZHVsZSgpCisgICAgICAgIHNlbGYuX3NhdmVfYW1peGVyX3N0YXRlKCkKIAog ICAgIGRlZiBjbGVhbl91cF90ZXN0X3J1bihzZWxmKToKICAgICAgICAgc2VsZi5fcmVzdG9yZV9w dWxzZWF1ZGlvX21vZHVsZSgpCisgICAgICAgIHNlbGYuX3Jlc3RvcmVfYW1peGVyX3N0YXRlKCkK IAogICAgIGRlZiBzZXR1cF9lbnZpcm9uX2Zvcl9zZXJ2ZXIoc2VsZiwgc2VydmVyX25hbWU9Tm9u ZSk6CiAgICAgICAgIGVudmlyb25tZW50ID0gV2ViS2l0UG9ydC5zZXR1cF9lbnZpcm9uX2Zvcl9z ZXJ2ZXIoc2VsZiwgc2VydmVyX25hbWUpCg== 163124 2012-09-10 07:14:55 -0700 2012-09-10 07:14:55 -0700 test volumechange event volumechange.html text/html 685 jussi.kukkonen PGh0bWwgbGFuZz0iZW4iPgo8Ym9keT4KICA8YXVkaW8gaWQ9ImEiIGNvbnRyb2xzPjwvYXVkaW8+ CiAgPGRpdiBpZD0ibG9nIj5Mb2c6IDwvZGl2PgogIDxzY3JpcHQ+CnZhciBhID0gZG9jdW1lbnQu Z2V0RWxlbWVudEJ5SWQoImEiKTsKdmFyIGxvZyAgPSBkb2N1bWVudC5nZXRFbGVtZW50QnlJZCgi bG9nIik7CgpmdW5jdGlvbiBjYW5wbGF5KCkgewogICAgbG9nLmlubmVySFRNTCArPSAiPGJyPiAq IGNhbnBsYXl0aHJvdWdoIGV2ZW50LCBtdXRlIGlzICIgKyBhLm11dGVkICsgIiwgdm9sdW1lIGlz ICIrIGEudm9sdW1lICsgIi4gTm93IGNhbGxpbmcgcGxheSgpIjsKICAgIGEucGxheSgpOwp9Cgpm dW5jdGlvbiB2b2x1bWVDaGFuZ2VkKCkgewogICAgbG9nLmlubmVySFRNTCArPSAiPGJyPiAqIHZv bHVtZWNoYW5nZSBldmVudCwgbXV0ZSBpcyAiICsgYS5tdXRlZCArICIsIHZvbHVtZSBpcyAiKyBh LnZvbHVtZTsKfQoKYS5hZGRFdmVudExpc3RlbmVyKCJjYW5wbGF5dGhyb3VnaCIsIGNhbnBsYXks IGZhbHNlKTsKYS5hZGRFdmVudExpc3RlbmVyKCJ2b2x1bWVjaGFuZ2UiLCB2b2x1bWVDaGFuZ2Vk LCBmYWxzZSk7CmEuc3JjPSJodHRwczovL3VwbG9hZC53aWtpbWVkaWEub3JnL3dpa2lwZWRpYS9j b21tb25zL2YvZjIvTWVkaWFuX3Rlc3Qub2dnIgogIDwvc2NyaXB0PgoKPC9ib2R5Pgo8L2h0bWw+ Cg== 432601 2021-06-30 09:01:11 -0700 2021-06-30 09:05:02 -0700 Patch bug-83704-20210630180110.patch text/plain 1617 calvaris U3VidmVyc2lvbiBSZXZpc2lvbjogMjc5NDA3CmRpZmYgLS1naXQgYS9Ub29scy9DaGFuZ2VMb2cg Yi9Ub29scy9DaGFuZ2VMb2cKaW5kZXggZWVjMjgwYWRmZDU2MGQ5YWY4MmY5MDAxNjkxZWZiMTFk MDAyNDMwMC4uODZiMzJiYzU1MTAzYjE1MDBjNmM4ZjljODY5YjcyZjc1YmZmMmZiMSAxMDA2NDQK LS0tIGEvVG9vbHMvQ2hhbmdlTG9nCisrKyBiL1Rvb2xzL0NoYW5nZUxvZwpAQCAtMSwzICsxLDE2 IEBACisyMDIxLTA2LTMwICBYYWJpZXIgUm9kcmlndWV6IENhbHZhciAgPGNhbHZhcmlzQGlnYWxp YS5jb20+CisKKyAgICAgICAgW0dUS10gbWVkaWEvZXZlbnQtYXR0cmlidXRlcy5odG1sIGZhaWxz IGlmIG1peGVyIGlzIG5vdCBhdCAxMDAlCisgICAgICAgIGh0dHBzOi8vYnVncy53ZWJraXQub3Jn L3Nob3dfYnVnLmNnaT9pZD04MzcwNAorCisgICAgICAgIFJldmlld2VkIGJ5IE5PQk9EWSAoT09Q UyEpLgorCisgICAgICAgICogV2ViS2l0VGVzdFJ1bm5lci9UZXN0Q29udHJvbGxlci5jcHA6Cisg ICAgICAgIChXVFI6OlRlc3RDb250cm9sbGVyOjpyZXNldFN0YXRlVG9Db25zaXN0ZW50VmFsdWVz KTogR1N0cmVhbWVyCisgICAgICAgIHBvcnRzIHVzZSBmYWtlc2luayB0byBhdm9pZCBzb3VuZCBv dXRwdXQgYW5kIHNldHRpbmcgdGhlIHBhZ2UKKyAgICAgICAgdm9sdW1lIHRvIDAgY3JlYXRlcyB0 cm91YmxlIHdpdGggdm9sdW1lIGV2ZW50cy4gTGV0J3MgYXZvaWQgaXQgaW4KKyAgICAgICAgR1N0 cmVhbWVyIHBvcnRzLgorCiAyMDIxLTA2LTI5ICBaaGlmZWkgRmFuZyAgPHpoaWZlaV9mYW5nQGFw cGxlLmNvbT4KIAogICAgICAgICBBZGQgbmV3IGFyZ3VtZW50IG9mIGRldmljZSBvcyB2ZXJzaW9u IGZvciByZXBvcnRpbmcgSmF2YVNjcmlwdENvcmUgdGVzdCByZXN1bHQKZGlmZiAtLWdpdCBhL1Rv b2xzL1dlYktpdFRlc3RSdW5uZXIvVGVzdENvbnRyb2xsZXIuY3BwIGIvVG9vbHMvV2ViS2l0VGVz dFJ1bm5lci9UZXN0Q29udHJvbGxlci5jcHAKaW5kZXggY2ZjOWM0NDgzYzgyNWQ5MTNlZDI3YTRk NTU4ODRmMjQyZDc2ZjAwMi4uNDM0MDlhYTA0NjJjMDZlZGIwMjVjYTRjMWM0YzNiODM4Y2QwYzIw NiAxMDA2NDQKLS0tIGEvVG9vbHMvV2ViS2l0VGVzdFJ1bm5lci9UZXN0Q29udHJvbGxlci5jcHAK KysrIGIvVG9vbHMvV2ViS2l0VGVzdFJ1bm5lci9UZXN0Q29udHJvbGxlci5jcHAKQEAgLTk4NCw3 ICs5ODQsMTAgQEAgYm9vbCBUZXN0Q29udHJvbGxlcjo6cmVzZXRTdGF0ZVRvQ29uc2lzdGVudFZh bHVlcyhjb25zdCBUZXN0T3B0aW9ucyYgb3B0aW9ucywgUmUKIAogICAgIFdLUGFnZUNsZWFyV2hl ZWxFdmVudFRlc3RNb25pdG9yKG1fbWFpbldlYlZpZXctPnBhZ2UoKSk7CiAKKyAgICAvLyBHU3Ry ZWFtZXIgdXNlcyBmYWtlc2luayB0byBhdm9pZCBzb3VuZCBvdXRwdXQgZHVyaW5nIHRlc3Rpbmcg YW5kIGRvaW5nIHRoaXMgY3JlYXRlcyB0cm91YmxlIHdpdGggdm9sdW1lIGV2ZW50cy4KKyNpZiAh VVNFKEdTVFJFQU1FUikKICAgICBXS1BhZ2VTZXRNZWRpYVZvbHVtZShtX21haW5XZWJWaWV3LT5w YWdlKCksIDApOworI2VuZGlmCiAKICAgICBXS1BhZ2VDbGVhclVzZXJNZWRpYVN0YXRlKG1fbWFp bldlYlZpZXctPnBhZ2UoKSk7CiAK 432602 2021-06-30 09:05:08 -0700 2021-07-01 08:08:13 -0700 Patch bug-83704-20210630180507.patch text/plain 3031 calvaris U3VidmVyc2lvbiBSZXZpc2lvbjogMjc5NDA3CmRpZmYgLS1naXQgYS9Ub29scy9DaGFuZ2VMb2cg Yi9Ub29scy9DaGFuZ2VMb2cKaW5kZXggZWVjMjgwYWRmZDU2MGQ5YWY4MmY5MDAxNjkxZWZiMTFk MDAyNDMwMC4uODZiMzJiYzU1MTAzYjE1MDBjNmM4ZjljODY5YjcyZjc1YmZmMmZiMSAxMDA2NDQK LS0tIGEvVG9vbHMvQ2hhbmdlTG9nCisrKyBiL1Rvb2xzL0NoYW5nZUxvZwpAQCAtMSwzICsxLDE2 IEBACisyMDIxLTA2LTMwICBYYWJpZXIgUm9kcmlndWV6IENhbHZhciAgPGNhbHZhcmlzQGlnYWxp YS5jb20+CisKKyAgICAgICAgW0dUS10gbWVkaWEvZXZlbnQtYXR0cmlidXRlcy5odG1sIGZhaWxz IGlmIG1peGVyIGlzIG5vdCBhdCAxMDAlCisgICAgICAgIGh0dHBzOi8vYnVncy53ZWJraXQub3Jn L3Nob3dfYnVnLmNnaT9pZD04MzcwNAorCisgICAgICAgIFJldmlld2VkIGJ5IE5PQk9EWSAoT09Q UyEpLgorCisgICAgICAgICogV2ViS2l0VGVzdFJ1bm5lci9UZXN0Q29udHJvbGxlci5jcHA6Cisg ICAgICAgIChXVFI6OlRlc3RDb250cm9sbGVyOjpyZXNldFN0YXRlVG9Db25zaXN0ZW50VmFsdWVz KTogR1N0cmVhbWVyCisgICAgICAgIHBvcnRzIHVzZSBmYWtlc2luayB0byBhdm9pZCBzb3VuZCBv dXRwdXQgYW5kIHNldHRpbmcgdGhlIHBhZ2UKKyAgICAgICAgdm9sdW1lIHRvIDAgY3JlYXRlcyB0 cm91YmxlIHdpdGggdm9sdW1lIGV2ZW50cy4gTGV0J3MgYXZvaWQgaXQgaW4KKyAgICAgICAgR1N0 cmVhbWVyIHBvcnRzLgorCiAyMDIxLTA2LTI5ICBaaGlmZWkgRmFuZyAgPHpoaWZlaV9mYW5nQGFw cGxlLmNvbT4KIAogICAgICAgICBBZGQgbmV3IGFyZ3VtZW50IG9mIGRldmljZSBvcyB2ZXJzaW9u IGZvciByZXBvcnRpbmcgSmF2YVNjcmlwdENvcmUgdGVzdCByZXN1bHQKZGlmZiAtLWdpdCBhL1Rv b2xzL1dlYktpdFRlc3RSdW5uZXIvVGVzdENvbnRyb2xsZXIuY3BwIGIvVG9vbHMvV2ViS2l0VGVz dFJ1bm5lci9UZXN0Q29udHJvbGxlci5jcHAKaW5kZXggY2ZjOWM0NDgzYzgyNWQ5MTNlZDI3YTRk NTU4ODRmMjQyZDc2ZjAwMi4uNDM0MDlhYTA0NjJjMDZlZGIwMjVjYTRjMWM0YzNiODM4Y2QwYzIw NiAxMDA2NDQKLS0tIGEvVG9vbHMvV2ViS2l0VGVzdFJ1bm5lci9UZXN0Q29udHJvbGxlci5jcHAK KysrIGIvVG9vbHMvV2ViS2l0VGVzdFJ1bm5lci9UZXN0Q29udHJvbGxlci5jcHAKQEAgLTk4NCw3 ICs5ODQsMTAgQEAgYm9vbCBUZXN0Q29udHJvbGxlcjo6cmVzZXRTdGF0ZVRvQ29uc2lzdGVudFZh bHVlcyhjb25zdCBUZXN0T3B0aW9ucyYgb3B0aW9ucywgUmUKIAogICAgIFdLUGFnZUNsZWFyV2hl ZWxFdmVudFRlc3RNb25pdG9yKG1fbWFpbldlYlZpZXctPnBhZ2UoKSk7CiAKKyAgICAvLyBHU3Ry ZWFtZXIgdXNlcyBmYWtlc2luayB0byBhdm9pZCBzb3VuZCBvdXRwdXQgZHVyaW5nIHRlc3Rpbmcg YW5kIGRvaW5nIHRoaXMgY3JlYXRlcyB0cm91YmxlIHdpdGggdm9sdW1lIGV2ZW50cy4KKyNpZiAh VVNFKEdTVFJFQU1FUikKICAgICBXS1BhZ2VTZXRNZWRpYVZvbHVtZShtX21haW5XZWJWaWV3LT5w YWdlKCksIDApOworI2VuZGlmCiAKICAgICBXS1BhZ2VDbGVhclVzZXJNZWRpYVN0YXRlKG1fbWFp bldlYlZpZXctPnBhZ2UoKSk7CiAKZGlmZiAtLWdpdCBhL0xheW91dFRlc3RzL0NoYW5nZUxvZyBi L0xheW91dFRlc3RzL0NoYW5nZUxvZwppbmRleCA1MzUyMzhkM2YwYWRjZDAzNThkNDRmMjZjYTg5 YjE4OTZmMGU1MWRmLi41ODYyZTk0NmNiNGMxYjk2NGJmODNmNmU5NTdiMTg3N2I2MTUwMjRhIDEw MDY0NAotLS0gYS9MYXlvdXRUZXN0cy9DaGFuZ2VMb2cKKysrIGIvTGF5b3V0VGVzdHMvQ2hhbmdl TG9nCkBAIC0xLDMgKzEsMTIgQEAKKzIwMjEtMDYtMzAgIFhhYmllciBSb2RyaWd1ZXogQ2FsdmFy ICA8Y2FsdmFyaXNAaWdhbGlhLmNvbT4KKworICAgICAgICBbR1RLXSBtZWRpYS9ldmVudC1hdHRy aWJ1dGVzLmh0bWwgZmFpbHMgaWYgbWl4ZXIgaXMgbm90IGF0IDEwMCUKKyAgICAgICAgaHR0cHM6 Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTgzNzA0CisKKyAgICAgICAgUmV2aWV3 ZWQgYnkgTk9CT0RZIChPT1BTISkuCisKKyAgICAgICAgKiBwbGF0Zm9ybS9nbGliL1Rlc3RFeHBl Y3RhdGlvbnM6IFVuZmxhZ2dlZCB0ZXN0LgorCiAyMDIxLTA2LTMwICBUaW0gTmd1eWVuICA8bnRp bUBhcHBsZS5jb20+CiAKICAgICAgICAgSW1wbGVtZW50IHN1cHBvcnQgZm9yIDxkaWFsb2c+IGVs ZW1lbnQgY2xvc2UgZXZlbnQKZGlmZiAtLWdpdCBhL0xheW91dFRlc3RzL3BsYXRmb3JtL2dsaWIv VGVzdEV4cGVjdGF0aW9ucyBiL0xheW91dFRlc3RzL3BsYXRmb3JtL2dsaWIvVGVzdEV4cGVjdGF0 aW9ucwppbmRleCA2OTZjMjViYTExYjc3OTRkMWQ5MzMyZjE5OTEyYjgxMDY5ODEwNjcwLi5lZDMz YmFhMjNmZGJlNDYxYmJkYTA2Nzg2YTE3NmE2Mjg5ZjQ1MWQ2IDEwMDY0NAotLS0gYS9MYXlvdXRU ZXN0cy9wbGF0Zm9ybS9nbGliL1Rlc3RFeHBlY3RhdGlvbnMKKysrIGIvTGF5b3V0VGVzdHMvcGxh dGZvcm0vZ2xpYi9UZXN0RXhwZWN0YXRpb25zCkBAIC0yMjE2LDcgKzIyMTYsNiBAQCB3ZWJraXQu b3JnL2IvMTg2Njc5IG1lZGlhL3ZpZGVvLWN1cnJlbnRUaW1lLWRlbGF5Lmh0bWwgWyBDcmFzaCBQ YXNzIFRpbWVvdXQgXQogCiB3ZWJraXQub3JnL2IvMTk4ODI3IG1lZGlhL21lZGlhLWZ1bGxzY3Jl ZW4tcmV0dXJuLXRvLWlubGluZS5odG1sIFsgVGltZW91dCBdCiAKLXdlYmtpdC5vcmcvYi8xOTg4 MzAgbWVkaWEvZXZlbnQtYXR0cmlidXRlcy5odG1sIFsgQ3Jhc2ggRmFpbHVyZSBdCiB3ZWJraXQu b3JnL2IvMTk4ODMwIG1lZGlhL3RyYWNrL3RyYWNrLWN1ZS1jc3MuaHRtbCBbIEltYWdlT25seUZh aWx1cmUgRmFpbHVyZSBQYXNzIF0KIHdlYmtpdC5vcmcvYi8xOTg4MzAgbWVkaWEvdmlkZW8tc2l6 ZS1pbnRyaW5zaWMtc2NhbGUuaHRtbCBbIEZhaWx1cmUgXQogd2Via2l0Lm9yZy9iLzE5ODgzMCBo dHRwL3Rlc3RzL21lZGlhL2F1dG9wbGF5LWlmLWF1ZGlvLWlzLXBsYXlpbmcuaHRtbCBbIFBhc3Mg VGltZW91dCBdCg== 432694 2021-07-01 08:08:17 -0700 2021-07-02 00:01:30 -0700 Patch bug-83704-20210701170816.patch text/plain 5155 calvaris U3VidmVyc2lvbiBSZXZpc2lvbjogMjc5NDA3CmRpZmYgLS1naXQgYS9Ub29scy9DaGFuZ2VMb2cg Yi9Ub29scy9DaGFuZ2VMb2cKaW5kZXggZWVjMjgwYWRmZDU2MGQ5YWY4MmY5MDAxNjkxZWZiMTFk MDAyNDMwMC4uMTNkMWVhNjg2ZWQ4YTg2MTEzNDYzMWI2M2YxMjk2MmE5YmYxODI1YyAxMDA2NDQK LS0tIGEvVG9vbHMvQ2hhbmdlTG9nCisrKyBiL1Rvb2xzL0NoYW5nZUxvZwpAQCAtMSwzICsxLDE2 IEBACisyMDIxLTA3LTAxICBYYWJpZXIgUm9kcmlndWV6IENhbHZhciAgPGNhbHZhcmlzQGlnYWxp YS5jb20+CisKKyAgICAgICAgW0dUS10gbWVkaWEvZXZlbnQtYXR0cmlidXRlcy5odG1sIGZhaWxz IGlmIG1peGVyIGlzIG5vdCBhdCAxMDAlCisgICAgICAgIGh0dHBzOi8vYnVncy53ZWJraXQub3Jn L3Nob3dfYnVnLmNnaT9pZD04MzcwNAorCisgICAgICAgIFJldmlld2VkIGJ5IE5PQk9EWSAoT09Q UyEpLgorCisgICAgICAgICogV2ViS2l0VGVzdFJ1bm5lci9UZXN0Q29udHJvbGxlci5jcHA6Cisg ICAgICAgIChXVFI6OlRlc3RDb250cm9sbGVyOjpyZXNldFN0YXRlVG9Db25zaXN0ZW50VmFsdWVz KTogR1N0cmVhbWVyCisgICAgICAgIHBvcnRzIHVzZSBmYWtlc2luayB0byBhdm9pZCBzb3VuZCBv dXRwdXQgYW5kIHNldHRpbmcgdGhlIHBhZ2UKKyAgICAgICAgdm9sdW1lIHRvIDAgY3JlYXRlcyB0 cm91YmxlIHdpdGggdm9sdW1lIGV2ZW50cy4gTGV0J3MgYXZvaWQgaXQgaW4KKyAgICAgICAgR1N0 cmVhbWVyIHBvcnRzLgorCiAyMDIxLTA2LTI5ICBaaGlmZWkgRmFuZyAgPHpoaWZlaV9mYW5nQGFw cGxlLmNvbT4KIAogICAgICAgICBBZGQgbmV3IGFyZ3VtZW50IG9mIGRldmljZSBvcyB2ZXJzaW9u IGZvciByZXBvcnRpbmcgSmF2YVNjcmlwdENvcmUgdGVzdCByZXN1bHQKZGlmZiAtLWdpdCBhL1Rv b2xzL1dlYktpdFRlc3RSdW5uZXIvVGVzdENvbnRyb2xsZXIuY3BwIGIvVG9vbHMvV2ViS2l0VGVz dFJ1bm5lci9UZXN0Q29udHJvbGxlci5jcHAKaW5kZXggY2ZjOWM0NDgzYzgyNWQ5MTNlZDI3YTRk NTU4ODRmMjQyZDc2ZjAwMi4uNDM0MDlhYTA0NjJjMDZlZGIwMjVjYTRjMWM0YzNiODM4Y2QwYzIw NiAxMDA2NDQKLS0tIGEvVG9vbHMvV2ViS2l0VGVzdFJ1bm5lci9UZXN0Q29udHJvbGxlci5jcHAK KysrIGIvVG9vbHMvV2ViS2l0VGVzdFJ1bm5lci9UZXN0Q29udHJvbGxlci5jcHAKQEAgLTk4NCw3 ICs5ODQsMTAgQEAgYm9vbCBUZXN0Q29udHJvbGxlcjo6cmVzZXRTdGF0ZVRvQ29uc2lzdGVudFZh bHVlcyhjb25zdCBUZXN0T3B0aW9ucyYgb3B0aW9ucywgUmUKIAogICAgIFdLUGFnZUNsZWFyV2hl ZWxFdmVudFRlc3RNb25pdG9yKG1fbWFpbldlYlZpZXctPnBhZ2UoKSk7CiAKKyAgICAvLyBHU3Ry ZWFtZXIgdXNlcyBmYWtlc2luayB0byBhdm9pZCBzb3VuZCBvdXRwdXQgZHVyaW5nIHRlc3Rpbmcg YW5kIGRvaW5nIHRoaXMgY3JlYXRlcyB0cm91YmxlIHdpdGggdm9sdW1lIGV2ZW50cy4KKyNpZiAh VVNFKEdTVFJFQU1FUikKICAgICBXS1BhZ2VTZXRNZWRpYVZvbHVtZShtX21haW5XZWJWaWV3LT5w YWdlKCksIDApOworI2VuZGlmCiAKICAgICBXS1BhZ2VDbGVhclVzZXJNZWRpYVN0YXRlKG1fbWFp bldlYlZpZXctPnBhZ2UoKSk7CiAKZGlmZiAtLWdpdCBhL0xheW91dFRlc3RzL0NoYW5nZUxvZyBi L0xheW91dFRlc3RzL0NoYW5nZUxvZwppbmRleCA1MzUyMzhkM2YwYWRjZDAzNThkNDRmMjZjYTg5 YjE4OTZmMGU1MWRmLi5jN2I0MWNlYzZlOTJiNTBkYmU0MTRiZDI5YmJlMDcyNTFjNDkxYTljIDEw MDY0NAotLS0gYS9MYXlvdXRUZXN0cy9DaGFuZ2VMb2cKKysrIGIvTGF5b3V0VGVzdHMvQ2hhbmdl TG9nCkBAIC0xLDMgKzEsMTMgQEAKKzIwMjEtMDctMDEgIFhhYmllciBSb2RyaWd1ZXogQ2FsdmFy ICA8Y2FsdmFyaXNAaWdhbGlhLmNvbT4KKworICAgICAgICBbR1RLXSBtZWRpYS9ldmVudC1hdHRy aWJ1dGVzLmh0bWwgZmFpbHMgaWYgbWl4ZXIgaXMgbm90IGF0IDEwMCUKKyAgICAgICAgaHR0cHM6 Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTgzNzA0CisKKyAgICAgICAgUmV2aWV3 ZWQgYnkgTk9CT0RZIChPT1BTISkuCisKKyAgICAgICAgKiBwbGF0Zm9ybS9nbGliL1Rlc3RFeHBl Y3RhdGlvbnM6IFVuZmxhZ2dlZCB0aGF0IHRlc3QgYW5kIG90aGVyCisgICAgICAgIHRoYXQgZ2V0 IGZpeGVkIHdpdGggdGhpcyBwYXRjaC4KKwogMjAyMS0wNi0zMCAgVGltIE5ndXllbiAgPG50aW1A YXBwbGUuY29tPgogCiAgICAgICAgIEltcGxlbWVudCBzdXBwb3J0IGZvciA8ZGlhbG9nPiBlbGVt ZW50IGNsb3NlIGV2ZW50CmRpZmYgLS1naXQgYS9MYXlvdXRUZXN0cy9wbGF0Zm9ybS9nbGliL1Rl c3RFeHBlY3RhdGlvbnMgYi9MYXlvdXRUZXN0cy9wbGF0Zm9ybS9nbGliL1Rlc3RFeHBlY3RhdGlv bnMKaW5kZXggNjk2YzI1YmExMWI3Nzk0ZDFkOTMzMmYxOTkxMmI4MTA2OTgxMDY3MC4uNDVkMTEy MGQxM2ZiY2E5NzY1YTBiNmY4OTljYTZmZTVlMGQ1ZDc1NCAxMDA2NDQKLS0tIGEvTGF5b3V0VGVz dHMvcGxhdGZvcm0vZ2xpYi9UZXN0RXhwZWN0YXRpb25zCisrKyBiL0xheW91dFRlc3RzL3BsYXRm b3JtL2dsaWIvVGVzdEV4cGVjdGF0aW9ucwpAQCAtMjE4Miw3ICsyMTgyLDYgQEAgd2Via2l0Lm9y Zy9iLzE2ODM3MyBtZWRpYS9tZWRpYS1wcmVsb2FkLW5vLWRlbGF5LWxvYWRldmVudC5odG1sIFsg RmFpbHVyZSBdCiB3ZWJraXQub3JnL2IvMTQzNzAyIG1lZGlhL3Byb2dyZXNzLWV2ZW50cy1nZW5l cmF0ZWQtY29ycmVjdGx5Lmh0bWwgWyBGYWlsdXJlIF0KIHdlYmtpdC5vcmcvYi8xOTg4MzAgbWVk aWEvY29udGV4dC1tZW51LWFjdGlvbnMuaHRtbCBbIFBhc3MgQ3Jhc2ggRmFpbHVyZSBdCiB3ZWJr aXQub3JnL2IvMTkzNjM4IG1lZGlhL3ZpZGVvLXdlYmtpdC1wbGF5c2lubGluZS5odG1sIFsgRmFp bHVyZSBdCi13ZWJraXQub3JnL2IvMjAxMjY2IG1lZGlhL2F1ZGlvLXBsYXliYWNrLXJlc3RyaWN0 aW9uLXBsYXktbXV0ZWQuaHRtbCBbIEZhaWx1cmUgXQogd2Via2l0Lm9yZy9iLzIyNDA2NyBtZWRp YS9tZWRpYS1zb3VyY2UvbWVkaWEtc291cmNlLXRpbWVzdGFtcG9mZnNldC10cmltLmh0bWwgWyBG YWlsdXJlIF0KIHdlYmtpdC5vcmcvYi8xNjAxMTkgbWVkaWEvdmlkZW8tb2JqZWN0LWZpdC1jaGFu Z2UuaHRtbCBbIFBhc3MgQ3Jhc2ggSW1hZ2VPbmx5RmFpbHVyZSBdCiAKQEAgLTIyMDgsNyArMjIw Nyw2IEBAIHdlYmtpdC5vcmcvYi8xNDM0NzggbWVkaWEvdmlkZW8tcGFnZS1sb2FkLXByZWxvYWQt bm9uZS5odG1sIFsgRmFpbHVyZSBdCiB3ZWJraXQub3JnL2IvMTQ1MDUxIHdlYmtpdC5vcmcvYi8x OTg4MzAgWyBSZWxlYXNlIF0gbWVkaWEvdmlkZW8tcnRsLmh0bWwgWyBJbWFnZU9ubHlGYWlsdXJl IFBhc3MgQ3Jhc2ggXQogCiB3ZWJraXQub3JnL2IvMTQ1MjU4IG1lZGlhL3ZpZGVvLWNvbnRyb2xs ZXItY2hpbGQtcmF0ZS5odG1sIFsgRmFpbHVyZSBQYXNzIF0KLXdlYmtpdC5vcmcvYi8xNTg5MTQg d2Via2l0Lm9yZy9iLzE5ODgzMCBtZWRpYS9yZXN0b3JlLWZyb20tcGFnZS1jYWNoZS5odG1sIFsg RmFpbHVyZSBdCiB3ZWJraXQub3JnL2IvMTE2Mjc3IG1lZGlhL3ZpZGVvLWJ1ZmZlcmVkLmh0bWwg WyBQYXNzIEZhaWx1cmUgIF0KIAogd2Via2l0Lm9yZy9iLzE4NjYzOCBtZWRpYS92aWRlby1yZW1v dGUtY29udHJvbC1wbGF5cGF1c2UuaHRtbCBbIEZhaWx1cmUgVGltZW91dCBQYXNzIF0KQEAgLTIy MTYsNyArMjIxNCw2IEBAIHdlYmtpdC5vcmcvYi8xODY2NzkgbWVkaWEvdmlkZW8tY3VycmVudFRp bWUtZGVsYXkuaHRtbCBbIENyYXNoIFBhc3MgVGltZW91dCBdCiAKIHdlYmtpdC5vcmcvYi8xOTg4 MjcgbWVkaWEvbWVkaWEtZnVsbHNjcmVlbi1yZXR1cm4tdG8taW5saW5lLmh0bWwgWyBUaW1lb3V0 IF0KIAotd2Via2l0Lm9yZy9iLzE5ODgzMCBtZWRpYS9ldmVudC1hdHRyaWJ1dGVzLmh0bWwgWyBD cmFzaCBGYWlsdXJlIF0KIHdlYmtpdC5vcmcvYi8xOTg4MzAgbWVkaWEvdHJhY2svdHJhY2stY3Vl LWNzcy5odG1sIFsgSW1hZ2VPbmx5RmFpbHVyZSBGYWlsdXJlIFBhc3MgXQogd2Via2l0Lm9yZy9i LzE5ODgzMCBtZWRpYS92aWRlby1zaXplLWludHJpbnNpYy1zY2FsZS5odG1sIFsgRmFpbHVyZSBd CiB3ZWJraXQub3JnL2IvMTk4ODMwIGh0dHAvdGVzdHMvbWVkaWEvYXV0b3BsYXktaWYtYXVkaW8t aXMtcGxheWluZy5odG1sIFsgUGFzcyBUaW1lb3V0IF0KQEAgLTIyMzEsMTYgKzIyMjgsMTAgQEAg d2Via2l0Lm9yZy9iLzE5ODgzMCBbIFJlbGVhc2UgXSBtZWRpYS92aWRlby1kZWZhdWx0cGxheWJh Y2tyYXRlLWJlZm9yZS1sb2FkLmh0bWwKICMgR0xYIGVycm9yIGluIGdzdF9nbF9jb250ZXh0X2ds eF9jcmVhdGVfY29udGV4dAogd2Via2l0Lm9yZy9iLzIxMDM3NCBodHRwL3Rlc3RzL21lZGlhL3Vz ZXItZ2VzdHVyZS1wcmVzZXJ2ZWQtYWNyb3NzLXhtbGh0dHByZXF1ZXN0Lmh0bWwgWyBGYWlsdXJl IFBhc3MgVGltZW91dCBdCiAKLXdlYmtpdC5vcmcvYi8yMDgyOTMgbWVkaWEvdmlkZW8tcmVzdHJp Y3RlZC1pbnZpc2libGUtYXV0b3BsYXktYWxsb3dlZC1pZi1hdWRpYmxlLmh0bWwgWyBGYWlsdXJl IF0KLQogaHR0cC90ZXN0cy9tZWRpYS92aWRlby1wcmVsb2FkLmh0bWwgWyBQYXNzIFNsb3cgXQog CiB3ZWJraXQub3JnL2IvMTA4OTI1IGh0dHAvdGVzdHMvbWVkaWEvdmlkZW8tcGxheS1zdGFsbC5o dG1sIFsgRmFpbHVyZSBUaW1lb3V0IF0KIAotd2Via2l0Lm9yZy9iLzEzMTU0NSBtZWRpYS9tZWRp YS1ldmVudC1saXN0ZW5lcnMuaHRtbCBbIFRpbWVvdXQgXQotd2Via2l0Lm9yZy9iLzEzMTU0NSBt ZWRpYS92aWRlby12b2x1bWUuaHRtbCBbIEZhaWx1cmUgXQotd2Via2l0Lm9yZy9iLzEzMTU0NSBt ZWRpYS9hdWRpby1wbGF5YmFjay12b2x1bWUtY2hhbmdlcy13aXRoLXJlc3RyaWN0aW9ucy5odG1s IFsgRmFpbHVyZSBdCi0KIHdlYmtpdC5vcmcvYi8xMzUwODYgbWVkaWEvdmlkZW8tc2Vlay13aXRo LW5lZ2F0aXZlLXBsYXliYWNrLmh0bWwgWyBUaW1lb3V0IFBhc3MgXQogd2Via2l0Lm9yZy9iLzEz NzY5OCBtZWRpYS92aWRlby1jb250cm9scy1kcmFnLmh0bWwgWyBUaW1lb3V0IF0KIHdlYmtpdC5v cmcvYi8xNDE5NTkgaHR0cC90ZXN0cy9tZWRpYS9jbGVhcmtleS9jbGVhci1rZXktaGxzLWFlczEy OC5odG1sIFsgQ3Jhc2ggVGltZW91dCBdCg==